app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Load model
model_name = "EleutherAI/pythia-1.4B-deduped"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Device
device = torch.device("cpu")
model.to(device)

def generate(prompt):
    inputs = tokenizer(prompt, return_tensors="pt").to(device)
    outputs = model.generate(
        **inputs,
        max_new_tokens=200,
        do_sample=True,
        temperature=0.7
    )
    text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return text

# Gradio interface làm API
iface = gr.Interface(
    fn=generate,
    inputs=gr.Textbox(lines=5, placeholder="Nhập prompt…"),
    outputs=gr.Textbox(),
    title="OASST-J-3B API"
)

iface.launch(server_name="0.0.0.0", server_port=7860)